The overall objective of the proposed research is to reduce the incidence of sexually transmitted infections (STIs) among college students. We propose to accomplish this by using the innovative, engineering-inspired multiphase optimization strategy (MOST) to develop a highly effective, appealing, economical, and readily scalable internet-delivered behavioral intervention targeting the intersection of alcohol use and sexual risk behavior. The rate of STIs on college campuses is alarming: one in four college students is diagnosed with an STI at least once during their college experience. Sexual activity when drinking alcohol is highly prevalent among college students. Alcohol use is known to contribute to the sexual risk behaviors that are most responsible for the transmission of STIs, namely unprotected sex, contact with numerous partners, and hook- ups (casual sexual encounters). Few interventions have been developed that explicitly target the intersection of alcohol use and sexual risk behaviors, and none have been optimized. In order to reduce the incidence of STI transmission among this and other high-risk groups, a new approach is needed. MOST is a comprehensive methodological framework that brings the power of engineering principles to bear on optimization of behavioral interventions. MOST enables researchers to experimentally test the individual components in an intervention to determine their effectiveness, indicating which components need to be revised and re-tested. Given the high rates of alcohol use and sex among college students, the college setting provides an ideal opportunity for intervening on alcohol use and sexual risk behaviors. The proposed study will include a diverse population of college students (? 50% African American) on 4 campuses: 2 Historically Black Colleges and Universities, 1 large public university, and 1 junior college. This will increase the generalizability of our findings. Our specific aims are to (1) develop and pilot test an initial set of online intervention components targeting the link between alcohol use and sexual risk behaviors, (2) use the MOST approach to build an optimized preventive intervention, and (3) evaluate the effectiveness of the newly optimized preventive intervention using a fully powered RCT. This work will result in a new, more potent behavioral intervention that will reduce the incidence of STIs among college students in the US, and will lay the groundwork for a new generation of highly effective STI prevention interventions aimed at other subpopulations at risk.